3 ms·
1) the zipped download has always the same name, becomes quickly confusing if you're grabbing several pages. Maybe adding a timestamp or, better, a title snippe
by vjandrea 10y ago
1) the zipped download has always the same name, becomes quickly confusing if you're grabbing several pages. Maybe adding a timestamp or, better, a title snippet could help.
2) found html tags and incorrect whitespace when exporting to TXT this page: https://www.packtpub.com/packt/offers/free-learning https://www.packtpub.com/packt/offers/free-learning
I want to try this tool with pages that Instapaper fails to grab.
- apancyborg 10y ago1) Understood, for the moment we have not yet decied which way we go but for sure we will stop giving only [documentcyborg.com].zip file 2) We found some bug in the export to rtf and txt (its getting fix), the other export are working correctly. For the Instataper pages that fail if you could send us some domain name you wish : hn at documentcyborg.com and we will test the parser against it to make sure it works.
- dredmorbius 10y agoFood for thought: 1. Externally-provided content is dangerous. You might use a hash of the domain name, but I'd avoid files named after the sources. 2. Metadata such as a date would be useful. 3. Despite 1, a highly-sanitised hostname could be informative. An iconv to 8-bit ASCII [-a-zA-Z0-9_], and not allowing the first character to be '-' might be a start. Put a length limit on that as well.
- apancyborg 10y agoThanks, for 1) this is based on extraction not source so very easy to mitigate, 2)right we can include it in the filename.3)we have include most of it in the new updated version online. Thank you very much for your report.
- apancyborg 10y agowe fixed all those point